2000 character limit reached
From Spectra to Biophysical Insights: End-to-End Learning with a Biased Radiative Transfer Model (2403.02922v1)
Published 5 Mar 2024 in cs.LG
Abstract: Advances in machine learning have boosted the use of Earth observation data for climate change research. Yet, the interpretability of machine-learned representations remains a challenge, particularly in understanding forests' biophysical reactions to climate change. Traditional methods in remote sensing that invert radiative transfer models (RTMs) to retrieve biophysical variables from spectral data often fail to account for biases inherent in the RTM, especially for complex forests. We propose to integrate RTMs into an auto-encoder architecture, creating an end-to-end learning approach. Our method not only corrects biases in RTMs but also outperforms traditional techniques for variable retrieval like neural network regression. Furthermore, our framework has potential generally for inverting biased physical models. The code is available on https://github.com/yihshe/ai-refined-rtm.git.
- Clement Atzberger. Development of an invertible forest reflectance model: The infor-model. In A decade of trans-european remote sensing cooperation. Proceedings of the 20th EARSeL Symposium Dresden, Germany, volume 14, pp. 39–44, 2000.
- Infogan: Interpretable representation learning by information maximizing generative adversarial nets. Advances in neural information processing systems, 29, 2016.
- The harmonized landsat and sentinel-2 surface reflectance data set. Remote sensing of environment, 219:145–161, 2018.
- Retrieval of canopy biophysical variables from bidirectional reflectance: Using prior information to solve the ill-posed inverse problem. Remote sensing of environment, 84(1):1–15, 2003.
- Active shape models-their training and application. Computer vision and image understanding, 61(1):38–59, 1995.
- A framework for the quantitative evaluation of disentangled representations. In International Conference on Learning Representations, 2018.
- Modeling radiative transfer in heterogeneous 3-d vegetation canopies. Remote sensing of environment, 58(2):131–156, 1996.
- Narendra S Goel. Models of vegetation canopy reflectance and their use in estimation of biophysical parameters from reflectance data. Remote sensing reviews, 4(1):1–212, 1988.
- P Gong. Inverting a canopy reflectance model using a neural network. International Journal of Remote Sensing, 20(1):111–122, 1999.
- Generative adversarial networks. Communications of the ACM, 63(11):139–144, 2020.
- Physics-informed machine learning: A survey on problems, methods and applications. arXiv preprint arXiv:2211.08064, 2022.
- beta-vae: Learning basic visual concepts with a constrained variational framework. In International conference on learning representations, 2017.
- Allometric equations for integrating remote sensing imagery into forest monitoring programmes. Global change biology, 23(1):177–190, 2017.
- A style-based generator architecture for generative adversarial networks. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, pp. 4401–4410, 2019.
- Auto-encoding variational bayes. arXiv preprint arXiv:1312.6114, 2013.
- Variational inference of disentangled latent concepts from unlabeled observations, 2018.
- Geometric-optical modeling of a conifer forest canopy. IEEE Transactions on Geoscience and Remote Sensing, (5):705–721, 1985.
- Challenging common assumptions in the unsupervised learning of disentangled representations, 2019.
- Opendr: An approximate differentiable renderer. In Computer Vision–ECCV 2014: 13th European Conference, Zurich, Switzerland, September 6-12, 2014, Proceedings, Part VII 13, pp. 154–169. Springer, 2014.
- Extracting interpretable physical parameters from spatiotemporal systems using unsupervised learning. Physical Review X, 10(3):031056, 2020.
- OpenAI. Gpt-4 technical report, 2023.
- A new forest light interaction model in support of forest monitoring. Remote Sensing of Environment, 42(1):23–41, 1992.
- Inversion of a forest reflectance model to estimate structural canopy variables from hyperspectral remote sensing data. Remote sensing of environment, 100(3):281–294, 2006.
- Gwynn H Suits. The calculation of the directional reflectance of a vegetative canopy. Remote Sensing of Environment, 2:117–125, 1971.
- Forest resilience, biodiversity, and climate change. In A synthesis of the biodiversity/resilience/stability relationship in forest ecosystems. Secretariat of the Convention on Biological Diversity, Montreal. Technical Series, volume 43, pp. 1–67, 2009.
- The fourth radiation transfer model intercomparison (rami-iv): Proficiency testing of canopy reflectance models with iso-13528. Journal of Geophysical Research: Atmospheres, 118(13):6869–6890, 2013.
- Physics-guided interpretable probabilistic representation learning for high resolution image time series. IEEE Transactions on Geoscience and Remote Sensing, 2022.
- Physics-constrained deep learning for biophysical parameter retrieval from sentinel-2 images: inversion of the prosail model. Available at SSRN 4671923, 2023.